Intel Shows How a CPU is Made

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

zerapio

Distinguished
Nov 4, 2002
396
0
18,780
[citation][nom]dicobalt[/nom]I guess I won't get an answer to those questions because the answers are what keeps intel in business lol[/citation]
Maybe if you start working there you'll get to know.
 

bin1127

Distinguished
Dec 5, 2008
736
0
18,980
Thanks TH and Intel for the article. The manufacturing process is pretty cool, especially the picture of the transistor interconnects. And I didn't know when they say tray pricing they meant literally a tray of processors. Good stuff!
 

anamaniac

Distinguished
Jan 7, 2009
2,447
0
19,790
Thank you so much Toms for bringing me this article, and thank you Tuan so writing it so well (apparantly I saw it after the typos were fixed).
Another thank you to Intel for publically sharing this. I love you Intel, and as long as I don't lose my job and can save up enough, I may be going i7/i5 for my next computer. =D

[citation][nom]JumpKickJoe[/nom]great now lets see them show us the process for their pricing strategies. :)[/citation]

This almost made me laugh and wake a few people up... how rude of you.

[citation][nom]iulianx[/nom]Great article! I had no idea how a cpu is made until now...[/citation]

Me neither. I ad guessed, but I was so horribly off... though my assumption was cooler.

[citation][nom]Wayoffbase[/nom]sand (earth), fire, water, and air. am I missing any? I don't think so you build the chip from earth, power it with fire magic, and tame the fire elemental inside with air or water if you are a really leet druid[/citation]

Way off base is right, +1 regardless. =D

[citation][nom]bin1127[/nom]Thanks TH and Intel for the article. The manufacturing process is pretty cool, especially the picture of the transistor interconnects. And I didn't know when they say tray pricing they meant literally a tray of processors. Good stuff![/citation]

Ha, I just thought like a box of processors...
How much is this silicon worth anyways?

[citation][nom]bloodblender[/nom]Saw this on APC's website, lol, pretty interesting stuff, but hey, at least Intel and AMD have got plenty of resources (sand) to keep making processors![/citation]

Ha. So, then, what would it take to do a massive die shrink and to move towards 1000mm wafers to save on costs even more though? Petaflop sever in every home anyone? Not like we're running out of materials anytime soon. Sand and copper are in a nice abundance these days (well copper somewhat).

[citation][nom]nonamelab[/nom]A little late Intel http://www.globalfoundries.com/mul [...] nd_to_chip[/citation]

I would like to thank you for the link, much appreciated.

[citation][nom]calmstateofmind[/nom]pretty interesting stuff. makes you wonder how somebody could actually think this stuff up and it work as well as it does. amazes me...[/citation]

Through a mixture of duct tape and bubble gum. Likely a lot of drunken ideas...

[citation][nom]JWL3[/nom]Anyone still think processors are overpriced?[/citation]

I never did. People paid several hundred dollars for a 1MHz chip during my fathers days, so this ain't so bad.

[citation][nom]dicobalt[/nom]That's an amazinly simple process to make a chip. I especially like how the multiple layers are made, that's so elegantly simple. The whole process looks like there is so much that could go wrong though. How do you polish a surface so smooth it's flat even on the atomic level? How do you prevent physical atomic level damage between manufacturing stages? How do you ensure contact between layers is perfect and there are no atomic sized airholes so to speak, is it perfect? Do I have little atomic sized vaccums inside my CPU between layers? Those are the most confusing parts to me. I guess I won't get an answer to those questions because the answers are what keeps intel in business lol[/citation]

We all have more questions. They would all takes between hours to even years to explain everything likely. Best bet is to become an engineer and apply to Intel for a high ranking position. =D
 
G

Guest

Guest
the first transistors made were huge, these cpu's are based on the same technology as those massive vaccume transistors that first started computing and such, only instead of being massive machines in their own right, we have simply shrunk them to fit millions and billions onto a tiny square piece of silicon.

Little green men aint got shit on us!!
 

omnimodis78

Distinguished
Oct 7, 2008
886
0
19,010
As nonamelab pointed out above, AMD beat Intel to the "press release" regarding the process (although it was never a real secret). Still, after watching this, and reflecting on the fact that humans are even capable of such things is inspiring. If we manage to resolve our current natural and social crises, I can only dream of the stuff that will be common place 200 years from now. Cheers!
 

kato128

Distinguished
Feb 23, 2009
158
3
18,685
[citation][nom]rvbeppler[/nom]"A microprocessor is the most complex manufactured product on earth."What about GPUs from ATi / nVidia?[/citation]

Technically a GPU is also a microprocessor.
 

socrates047

Distinguished
Apr 23, 2009
175
0
18,680
they dont say what type of cooling they use during binning. do they use the stock intel coolers or big aftermarket kinds? or even water?

there should be a way to find out the chip's headroom straight from the manufacturer.. im just saying. great article!
 

scook9

Distinguished
Oct 16, 2008
826
0
18,980
cool article, emailing to my linear electronics professor now (that is where we learn what a transistor is and how to do circuit applications with it)
 

flinxsl

Distinguished
Sep 5, 2008
42
0
18,530
wow this is actually a pretty complete article, GJ intel. the hardest part by far is to achieve planarization between layers (making everything flat). This limits the place where I work to having only two metal layers, where intel has like 9+, lol.

if anyone is interested heres the lecture notes for the first semiconductor class i took on MOS processing, a little dated, but the teacher used to work for intel in folsom. http://gaia.ecs.csus.edu/~pheedley/CpE151/cpe151_lec4_mos_processing.pdf
 

Titanius

Distinguished
Sep 24, 2007
118
0
18,690
The copper ions travel from the positive terminal (anode) to the negative terminal (cathode) which is represented by the wafer.

That is wrong. It should be:

The copper ions travel from the positive terminal (cathode) to the negative terminal (anode) which is represented by the wafer.
 
Status
Not open for further replies.